Root Cause Analysis for Radiation Oncology. Steven Sutlief, PhD Department of Veterans Affairs

Transcription

1 Root Cause Analysis for Radiation Oncology Steven Sutlief, PhD Department of Veterans Affairs 1

2 Acknowledgements I want to express my thanks to Ed Leidholdt, Jr., Carl Bergsagel, the VA Radiation Oncology Field Advisory Group, and my fellow members of the AAPM Working Group for the Prevention of Medical Errors for insightful discussion of error prevention. Conflict of Interest: None. Mention of commercial products does not constitute endorsement of those products. 2

3 Objectives Understanding RCA outside/within healthcare Determining appropriate uses for RCA Knowing tools for performing RCA Assessing role of RCA in error prevention 3

4 What is Root Cause Analysis? 4

5 Introduction: Keys Parts of RCA Start with a problem whose causes may not be obvious. Get the right people on the analysis team. Visit the site, collect data, interview participants Create an event sequence diagram. Identify causes and effects, then keep asking why ( 5 times ) to find what is under each cause. Formally report recommendations and track the efficacy of the corrective actions. 5

6 Root Cause Analysis Outside and Inside Healthcare 6

7 Understanding RCA outside Healthcare RCA started in engineering (1960s) and was exported to many disciplines: Methodology Origin Application Safety-based Accident analysis Occupational Safety and Health. Production-based Quality control Industrial Manufacturing. Process-based Production-based Business processes. RCA Failure-based Failure analysis Engineering and Maintenance. Systems-based Amalgamation of the preceding schools 7

8 Understanding RCA within Healthcare The Joint Commission established a requirement for RCA in its 1999 sentinel event standards. The Department of Veterans Affairs National Center for Patient Safety was established in Institute of Medicine issued To Err is Human in 1999, followed by multiple reports on error reduction, including Patient Safety in (Charles Perrow published Normal Accidents: Living with High-Risk Technologies in 1999.) 8

9 VHA Event Dissemination The VA National Health Physics Program has provided newsletters describing misadministrations. The VA s reporting requirement for radiotherapy medical events included an RCA process which produced a set of lessons learned to be shared in the newsletters along with an anonymized description of the event. 9

10 Understanding RCA within Healthcare The Joint Commission encourages, but does not require, self-reporting of sentinel events. If the Joint Commission becomes aware of a sentinel event, the accredited organization is expected to prepare a thorough and credible root cause analysis and action plan. Sentinel event is defined on one of the later slides. 10

11 When to Perform RCA? When Mandated Or When Warranted by Interaction/Coupling 11

12 When to Perform RCA? When required to by an oversight agency: US Nuclear Regulatory Commission The Joint Commission When required by Medical Center policy. When deemed worth the time and effort to perform more than a superficial analysis: Taxonomies are being developed to assist in how we capture events and categorize them. 12

13 Interaction/Coupling Diagram Linear INTERACTIONS Complex Tight Dams Power Grids Aircraft Nuclear plant Rail Transport Space missions Loose COUPLING Assembly-line production Most Manufacturing Military Adventures Mining R & D Firms 7/21/2010 Adapted from Normal Accidents, Charles Perrow, p97 Steven Sutlief, AAPM

14 Interaction/Coupling for Rad Onc Tight Loose COUPLING Linear Patient Simulation Patient plan hand off to the therapists Daily Linac Quality Assurance Monthly Linac Quality Assurance INTERACTIONS Complex Stereotactic Radiosurgery Plan and Treat Rad Onc Data communications maintenance Annual Linac Quality Assurance IMRT Treatment planning and assessment RadOnc Patient Admission and Discharge. Patient Setup 14

15 Chronic vs Sporadic Events Sporadic problem: a sudden adverse change in the status quo, requiring remedy through restoring the status quo. Example: Patient receives multiple fractions before staff discover the iso-shift was incorrect. Chronic problem: a long-standing adverse situation, requiring remedy through changing the status quo. Example: Physician routinely fails to review portal imaging before first treatment fraction, leading to occasional delays in treatment. 15

16 How does one perform RCA? 16

17 Simple Framework for RCA Chronological sequence Diagram the flow of events leading up to the incident (including the three whys ) Cause and Effect Diagramming Identify the conditions that resulted in the adverse event or close call Causal Statements Develop root cause and contributing factor statements, actions, and outcomes 17

18 The Three Whys When distilling the event narrative into an event flow diagram, it is useful to ask the three whys: What happened? Why did it happen? What are you going to do about it? 18

19 Radiation Therapy Example Initial event flow diagramming: use the event narrative to construct the discrete events in chronological order. Patient undergoes radiotherapy One day the therapists switch the beam order to facilitate filming Second field is erroneously delivered at location of first field Patient receives excess radiation at the location of the first field 19

20 Radiation Therapy Example Intermediate event flow diagramming: ask why each event occurred until there are either no more questions or no more answers. Patient undergoes radiotherapy One day the therapists switch the beam order Second field is erroneously treated at location of first field Patient receives excess radiation at the location of the first field Question: Why was switching field order done? Answer: Reordering the fields simplified filming of the field portals. Question: Why wasn t normal setup followed? Answer: The physicist interrupted the therapist with a question. Question: Why wasn t the field position verified? Answer: No second therapist was available. 20

21 Radiation Therapy Example Final event flow diagramming: done after answering why questions, interviews, and reference review. Patient undergoes radiotherapy Therapists realize portal imaging would be faster if fields reordered One day the therapists switch the beam order Physicist distracts therapist with a question during patient setup Subsequent tx modified to account for excess dose Patient receives excess radiation at the location of the first field No 2nd therapist available to double check field setup Second field is erroneously treated at location of first field 21

22 Radiation Therapy Example Cause and Effect Diagramming: Review the event flow diagram and clarify the problem statement. Brainstorm a list of causes and choose the most important. Complete the causal chain. Conclude the investigation by developing root cause and contributing factors statements. Patient receives excess radiation (Problem Statement) 22

23 Radiation Therapy Example Cause/Effect Diagramming Brainstorming: Patient receives excess radiation (Problem Statement) Causal chain: Distracted Therapist doesn t fix setup (Action) Interruption by physicist No 2 nd therapist to verify fields Distracted Therapist doesn t fix setup (Action) Rush to get back on schedule (Condition) caused by caused by Interruption by physicist (Condition) No 2 nd therapist to verify fields (Condition) Lack of other staff to ask 2 nd therapist pulled elsewhere 23

24 Radiation Therapy Example Root Cause/Contributing factor statements -- The Five Rules of Causation: Clearly show the cause and effect relationship. Use specific descriptors, not vague words. Identify preceding causes, not human error. Identify preceding causes of procedure violations. Failure to act is only casual when there is a preexisting duty to act. 24

25 Radiation Therapy Example Root Cause/Contributing factor statements: Therapist was distracted by other staff. Complex field arrangement. Therapists didn t perform independent checks. Root: procedures did not prohibit reversing field order or require independent field checks. 25

26 Radiation Therapy Example Actions: Two therapists will identify each field prior to delivery. The field order of complex field arrangements will not be reversed for convenience. Staff training for revised policies. Outcome measures: Tracking of future setup errors. 26

27 RCA Implementation Post-it Notes Presentation Board Software 27

28 Useful tools The Joint Commission s Framework for Conducting a Root Cause Analysis and Action Plan is an excellent six-page Word document that leads you through the steps of RCA. The VA National Center for Patient Safety Root Cause Analysis Tools flip book is an excellent tool for performing RCA as described in this talk. 28

29 Other Tools (Commercial) PROACT : Methodology and software for performing RCA, FMEA, and Opportunity Analysis. Reliability Center Inc. REASON : Training and software for performing RCA. Decision Systems, Inc. TapRooT : Training and software for the incident investigation process including RCA. System Improvements, Inc. Others 29

30 Available Reports Containing RCA The Radiation Therapy Incident at the Centre Hospitalier Jean Monet, Epinal, France Report into unintended overexposure of Lisa Norris at Beatson, Glasgow Treatment mistakenly delivered without wedge for 14 of 15 treatment fractions Error in transfer of treatment plan from R&V to linac, MLC fully open for three treatment fractions Error in commissioning orthovoltage machine, 620 patients treated, no independent check References are given in the slide-notes. (References are given in the slide-notes.) 30

31 More Reports Containing RCA IAEA reports: Accidental Overexposure of Radiotherapy Patients in Bialystok Investigation of an Accidental Overexposure of Patients in Panama Accidental Overexposure of Radiotherapy Patients in San Jose, Costa Rica 31

32 Conclusions Root Cause Analysis may be prompted by an outside agency, the medical center, or internal to the department. One methodology for RCA consists of: Flow diagramming Cause and effect diagramming Causal statements, and Actions and outcomes Everyone has their own formalism for RCA. Pick one. 32

33 References Error Reduction in Health Care, Patrice L. Spath, Ed., Patient Safety: Achieving a New Standard for Care, Institute of Medicine, Patient Safety: The PROACT Root Cause Analysis Approach, Robert J. Latino, Understanding Patient Safety, Robert M. Wachter,

34 References (continued) Normal Accidents: Living with High-Risk Technologies, Charles Perrow, To Err is Human, Institute of Medicine, Crossing the Quality Chasm, Institute of Medicine, TJC, Sentinel Event Policy and Procedures, 2004, available on the Web. 34

35 References (continued) Techniques for Root Cause Analysis, Patricia M. Williams, Proc (Bayl Univ Med Cent) April; 14(2): Holmberg O, McClean B. Preventing treatment errors in radiotherapy by identifying and evaluating near misses and actual incidents. J Rad Ther Practice 3:13-25 (2002) ICRP Publication 86: Prevention of Accidents to Patients Undergoing Radiation Therapy 35

36 References (end) IAEA - Prevention of Accidental Exposure in Radiotherapy (slides) IAEA - Accident Prevention (html) IAEA - Lessons Learned from Accidental Exposures in Radiotherapy, IAEA Safety Reports Series No. 17, IAEA, Vienna (2000) 36